Information for site owners
PukalaniMarketBot
If this name appears in your access log, our market comparison has read a public page of your website. This page explains what happens — and how to stop it.
User-agent: PukalaniMarketBot/1.0 (+https://branding.supply/market-bot)What it reads
Publicly reachable marketing pages only: home, about, services, pricing and the like. They are found through your home page, your sitemap and, if present, your llms.txt.
- —At most 8 pages per website and run.
- —At most 20,000 characters of text and 2 MB per page.
- —No browser, no JavaScript, no forms, no downloads — only the HTML you serve.
- —One page after another, never in parallel. Fetching is a read, not a load test.
What is never requested
Pages carrying these path segments are skipped before any request is made — personal data, legal texts, functional pages and dated content:
team · teams · ueber-das-team · mitarbeiter · people · staff · impressum · imprint · legal-notice · kontakt · contact · kontaktformular · jobs · job · karriere · career · careers · stellenangebote · presse · press · pressemitteilungen · datenschutz · privacy · privacy-policy · datenschutzerklaerung · agb · terms · terms-of-service · nutzungsbedingungen · widerruf · cookies · cookie-policy · barrierefreiheit · accessibility · login · signin · anmelden · register · registrieren · signup · account · konto · profil · profile · dashboard · cart · warenkorb · checkout · kasse · zahlung · search · suche · sitemap · feed · rss · blog · news · neuigkeiten · aktuelles · magazin · journal · events
From the remaining text, e-mail addresses, phone numbers and personal names are removed before any language model sees it. The bot never touches login areas and never signs in anywhere.
What happens to what was read
- —Page text is cleared after at most 24 hours. It is an intermediate product, not a record.
- —What remains are short quotes of at most 200 characters, each with the address of the page it stands on.
- —No ranking and no rating of your brand is produced. We show what brands say — not how successful they are with it.
- —No advertising line naming you or making you identifiable is derived from the result.
- —None of it is used to train models. The providers we use run without data retention.
How to shut us out
Two ways, both effective on the next run. Being excluded means your website is not analysed, and the customer who started the comparison is shown exactly that as the reason.
1. robots.txt
We fetch your robots.txt before every access and obey it — including rules addressed to all bots.
User-agent: PukalaniMarketBot
Disallow: /2. Usage reservation (text and data mining)
We recognise a machine-readable reservation under EU DSM Article 4 in these four forms. Any one of them is enough:
As an HTTP header
TDM-Reservation: 1As an HTML meta tag
<meta name="tdm-reservation" content="1">As a robots meta tag
<meta name="robots" content="noai, noimageai">As tdmrep.json under /.well-known/
/.well-known/tdmrep.json [{ "location": "/", "tdm-reservation": 1 }]
In case of doubt we decide against analysis: if a reservation cannot be safely ruled out, your website is excluded.
Questions, objection, deletion
Write to us. On request we exclude your domain permanently, without you having to change anything on your website.